Gold has the property, unique among the elements, of causing otherwise sensible people to lose all reason. Among the latest casualties are Shao Fenggao, an executive with China Construction Bank Corp, and Meng Qingfa, an analyst at the China Chamber of International Commerce. Worried that the Federal Reserve's policy of quantitative easing will undermine the value of the US dollar, both gentlemen have gone on the record in the past couple of weeks to argue that Beijing should increase its holdings of gold to diversify China's foreign exchange reserves away from the US currency. Now, on first hearing this might sound like a smart idea. After all, while the US dollar has slumped 35 per cent against a basket of currencies in the past 10 years, the price of gold has quadrupled to US$1,361 an ounce. So gold looks like a great hedge against US dollar weakness. The only problem is that the hedge doesn't work for the People's Bank of China. At the end of September, the Chinese central bank was sitting on foreign reserves worth US$2.65 trillion. Most of that stash - probably around two-thirds - is held in US dollar-denominated assets, although according to the latest available figures, China's reserves also include 1,054 tonnes of gold. At end-of-September prices, that holding would have been worth US$44.4 billion, making up 1.7 per cent of China's reserves. To achieve any meaningful diversification, Beijing would have to hold at least 5 per cent of its reserves in gold, which means - again at September's prices - China's reserve managers would have to buy another 2,087 tonnes of gold. This is clearly a big load of bullion. To put it into perspective, last year all the gold mines in the world managed to supply only 2,322 tonnes of the metal. As a result, for China to diversify into gold by buying bullion on the open market would hardly be practical. At the first hint of buying interest from the PBOC, the gold price would soar to stratospheric heights. Granted, that would boost the value of the gold it already holds, raising the proportion of gold by value in the central bank's portfolio, but the diversification benefits would be meaningless. Buying from the market is not a realistic solution. Alternatively, the PBOC could try to buy off-market, which would mean purchasing bullion from another central bank willing to sell. That would make more sense, but even so buying 2,000 tonnes of bullion would not be simple. Only a handful of central banks hold so much bullion (see the chart below). And although many reserve managers these days are dubious about the merit of gold as a reserve asset, central banks are extremely sensitive to the criticism they would bring on themselves if they sold ahead of further price gains. In any case, it is doubtful that Beijing would be an eager buyer of gold at current prices. A year ago, China had the opportunity to buy 200 tonnes of gold from the International Monetary Fund, but the PBOC baulked at paying the market price, which was then around US$1,060 or some 20 per cent below the current rate. But even if Beijing did manage to buy the 2,000-odd tonnes of gold it would need to achieve even a minimal diversification of its reserves, there would still be a problem. China's reserves are not static. For the past five years, they have grown at an astonishing average rate of US$31 billion a month. That means at current prices, China's reserve managers would have to buy more than 35 tonnes of gold a month just to maintain their weighting to gold. That equates to 425 tonnes a year or nearly 20 per cent of the total supply from all the world's mines. Alas, diversifying into gold simply isn't feasible. Speaking of gold, the Hong Kong Mercantile Exchange still hopes to go live with a futures contract on the metal before the end of the year. The new contract has been a long time in gestation. HKMEx was set up in June 2008, but has yet to start trading. Whether the wait will prove worthwhile is doubtful. The launch period has been so protracted that in the interval Singapore has managed to establish its own commodity exchange, the Singapore Mercantile Exchange, which began trading its own gold futures contract at the end of August. The results have been disappointing. In October, the leading gold future traded just 554 contracts. At the end of the month, open interest - the key measure of investor commitment - in gold futures was just 95 lots, worth a notional US$4 million. In contrast, New York's Comex exchange, which offers round-the-clock electronic dealing in gold futures to Asian investors, traded 3.86 million contracts in October. Month-end open interest was 620,000 lots, worth a notional US$83 billion. If Singapore has done so poorly at drawing trading volumes away from the established market, it is hard to see how HKMEx, as a Johnny-come-very-lately to the market, will be able to do any better.